Prompt Database:
Save multiple prompts (or whatever) with names, in three different categories (three separate sets of saved prompts). It's stored in the node's folder, in json file. so all saved prompts are available in any workflow in your comfy installation. Should be simple to understand.
Of course it doesn't have to be prompts. You can use it as simple database for whatever, any name and any text. Keep in mind it is saved in plain text in the nodes local folder! (a5prompt_database.json)
LMStudio Prompt Enhancer:
Send any system prompt to a localhost server (Example LMStudio) and/or a user prompt (to improve) and/or a image. (requires a vision model oc) and get the llm's answer back, exportable as string for any text input in Comfyui. See community post (or node tooltips!) for some important details. Three modes "Always run LLM", "Use last prompt" (which can be manually edited!) and "Bypass if unchanged", also sends the last prompt, if inputs and settings have not changed., But runs the LLM again IF something has changed. Optional load/unload model for generation, dropdown for last 5 used models, Api token, max tokens, auto removes thinking blocks. SEE community post on how to get the modelID from LMstudio and add a model to the list.
Installation: copy/clone directories or extract zip into Comfyui Custom_nodes See community post for more details, images. See example Workflow, using both nodes, under files. NOT on github/comfyui Manager
They dynamic Duo at work (this is the example workflow under files, i recommend adding a clear vram node Before it, so Comfyui kicks any loaed model out of vram, before LM Studio loads it's model):
Added notes: To use the load and unload model switches, LMStudio (not me) requires a API token to be set.
But you don't need a token to just send it to whatever model the LMStudio server has currently loaded. Just disable the load/unload switches and it's in easy mode. the "require authentication" option needs to be disabled in LMStudios server settings of course (else you need a api token!).
Also i'd recommend putting a "clear vram" node in front of it, so it kicks any comfy models out of vram before loading the lm studio model! Else LMStudio may try to load that xxGB LLM model while LTX or whatever is still in it.
How to copy a model id to add to the node? :
remember you can leave it empty, leave load/unload unselected and so simply send it to whatever is loaded in the LMStudio Server!
Like if you like :)